Skip to main content

Table 6 Scenario 1: perplexities of the adapted linguistic models. The baseline system (linguistic models are trained on Ohm_N) is given in column "Ohm_N." "" denotes the arithmetic (unweighted) mean. The average of the four subsets weighted by the prior probabilities of the four classes is given in row "Mont."

From: On the Impact of Children's Emotional Speech on Acoustic and Language Models

 

Linguistic models trained on

Test set

Ohm_M

Ohm_N

Ohm_E

Ohm_A

Mont_M

27.2

39.2

87.4

74.5

Mont_N

38.4

20.7

35.8

30.3

Mont_E

31.4

13.2

9.93

12.4

Mont_A

24.7

12.6

12.3

9.05

30.4

21.4

36.4

31.6

Mont

36.7

19.7

31.0

26.9